The study’s objective was to evaluate pregnancy, delivery and neonatal outcomes in multifetal gestations conceived through IVF compared to naturally conceived multifetal gestations based on a large population-sized American database.
In this study, we have found that IVF multiple gestation pregnancies were more likely complicated by PIH, gHTN, PET, GDM and placenta previa. Additionally, women who underwent IVF were more likely to be older which makes initiative sense because this group is more likely to be infertile and have multiple embryos transferred. Both increasing maternal age and IVF are reported to be linked with increased risk for pregnancy induced hypertensive disorders. We controlled for age in our analysis and still found this risk to be increased. This augmented risk of hypertensive disorders during pregnancy was also found in singleton pregnancies from IVF as shown by Pandey and coauthors in their meta-analysis and systematic review.[27]
This increased risk may be due to the hyperestrogenic state induced during IVF associated with endothelial dysfunction.[28] This dysfunction can lead to abnormal placentation and an increased risk of hypertensive disorders. It is theorized that the hormonal milieu during IVF compared to that in spontaneous pregnancy along with the preexisting underlying metabolic-vascular state of patients undergoing IVF may play a role in the development of GDM and PIH. [28] Embryo transfer and embryo culture affect implantation and embryo development; thus, abnormal placentation such as placenta previa can occur. This is hypothesized to be a result of uterine contractions caused by the embryo transfer via the trans-cervical catheter; which can result in mechanical stimulation of the internal os and thus release of prostaglandins. [29,30]
The risk of PPH is increased in multifetal pregnancies.[31] Our study has shown that multifetal pregnancies resulting from IVF are at increased risk of PPH when compared to spontaneous multiple gestations. This increased risk has been shown previously to exist with singleton pregnancies resulting from IVF. [27]
Congenital anomalies were also more likely to occur and probably share the same pathophysiology as singleton pregnancy through IVF. A meta-analysis comparing rates of birth defects in singleton pregnancies conceived through ART vs. natural conception found that neonates born after ART had a higher risk of birth defects.[23] They postulated that this may be due to characteristics of the infertile couple or the process of IVF/ICSI such as ovarian hyperstimulation, media culture of the embryo, and the freezing and thawing process of the embryo which may affect embryo differentiation through altered methylation and gene expression.[31] Even though multiple pregnancies are known risk factors for congenital anomalies [32] it appears that the use of IVF is additive to this risk, raising the risk substantially by about 80%.
While the literature is limited on delivery outcomes such as preterm delivery and SGA; studies have demonstrated an increased risk of these complications in pregnancies achieved via IVF. [24,26] This increased risk could be hypothesized to be attributed to the IVF cycle characteristic itself or the inherent features of patients who require IVF to conceive. [24, 25] Again, multiple pregnancies are known to be risks for preterm delivery and SGA. We however found that IVF conceived multiple pregnancies are further at risk for these complications with the risk increased by 7% and 26% respectively.[33] Studies have shown that singleton pregnancies conceived through IVF also have an increased risk of SGA when compared to naturally conceived singleton pregnancies. [26,27,28] Furthermore, increased levels of insulin-like growth factor-binding protein levels found in ART pregnancies have been linked to intrauterine growth restriction. [29]
A meta-analysis comparing risks of spontaneous preterm births in singleton pregnancies conceived with IVF or ICSI vs. spontaneously found an increased risk of 80% in pregnancies conceived through IVF or ICSI.[34] It is also important to keep in mind iatrogenic causes of preterm birth in IVF pregnancies due to abnormal placentation and in some cases may be related to the patient. [27,28,35] On the other hand, we found no statistical difference in preterm delivery rates. In our study, delivery outcomes in IVF multifetal gestations were more likely to be complicated by PPROM, chorioamnionitis, PPH, transfusion, maternal infection and CS as compared to spontaneously conceived multiple gestations. These findings remain consistent with published results seen in IVF singleton pregnancies.[36] It should be noted that in our study there were no statistical differences between the two groups for the outcomes: abruptio placenta rates, operative vaginal delivery, hysterectomy and maternal death. Some of which is due to the small rate of incidence of these complications; for example, peripartum hysterectomy complicates 1 per 1,000 deliveries in the United States. However, in the literature, IVF was found to have an increased risk of preterm delivery and perinatal mortality in singleton pregnancies, when meta-analyses are performed. [37,38] Singleton IVF pregnancies demonstrate a higher risk associated with IVF with regards to the associated risk of hysterectomy. [39] This is in contrast to our findings in multifetal IVF pregnancies and is likely related to the increased risks of hysterectomy deliveries in multifetal gestations overshadowing the risk of IVF. The same explanation can be applied to the lack of risk difference with regards to abruptio placenta as a complication [40,41]
The increased risk of CS seen in IVF multiple gestations may be related to requested CS and some providers being hesitant to attempt vaginal delivery in women with advanced maternal age, twins and IVF. This is similar to findings of other studies looking at multiple gestation IVF pregnancies. [7] It is important that future studies attempt to address the cause of this difference.
Contrary to our findings of the increased risk of maternal hypertension in IVF multifetal pregnancies; a Dutch study by Szymusik et al, comparing a cohort of similar European patients found no increased risk between IVF multiples and spontaneous multiple pregnancies.[42] It is likely this difference in finding in our study is due to the different populations studied and the increased risk of hypertension seen in our study may be related to the American population. Additionally, where we found an increased risk of SGA in IVF twin pregnancies, the Dutch study by Szymusik et al found no increased risk of SGA. It is worth mentioning however that in a subsequent meta-analysis the relative risks of SGA were similar in IVF versus spontaneously conceived multiple pregnancies. [43] The role of geographic location on this finding should also be considered as contributing to the outcomes differences seen.
There are several limitations to this study. To begin with, the reliance on a retrospective database is a known risk for coding errors and undetermined biases, this is a limitation of all large population databases. However, we prefer such types of studies, given the large number of subjects that could be included. In our case, this was approximately 100,000 multiple gestations. Information about infertile subjects with spontaneous conceptions wasn’t available in this database, making it impossible to determine whether pregnancy complications are associated with the IVF treatments as opposed to underlying infertility. Moreover, the database does not permit separation of frozen embryo transfers and fresh embryo transfers which may have given slightly different results and types of complications. Furthermore, it is likely the use of IVF is under-represented in the database and that some of the multiple gestations that acted as the controls may have had IVF. However, this would only serve to minimize differences between the groups and as such any increases in pregnancy complications in the IVF group are likely true.
Some of the main strengths of this study are that it is the first of its kind in North America in addition to the inclusion of such a large number of multiple gestations.